Andreas Madsen
@andreasmadsen.bsky.social
Ph.D. in NLP Interpretability from Mila. Previously: independent researcher, freelancer in ML, and Node.js core developer.
Also thanks to @sarath-chandar.bsky.social and @sivareddyg.bsky.social for supporting me during my Ph.D., which helped me get this far! I would highly recommend them if you are looking for a Ph.D. supervisor.
February 7, 2025 at 5:01 PM
Also thanks to @sarath-chandar.bsky.social and @sivareddyg.bsky.social for supporting me during my Ph.D., which helped me get this far! I would highly recommend them if you are looking for a Ph.D. supervisor.
Positions:
* Full-stack
* Research Engineer
* Research Scientist
* Systems Infrastructure Engineer
* Research intern
Feel free to reach out but chances are I will see your application if you apply online. I will post details on my internship later, but there are more openings.
* Full-stack
* Research Engineer
* Research Scientist
* Systems Infrastructure Engineer
* Research intern
Feel free to reach out but chances are I will see your application if you apply online. I will post details on my internship later, but there are more openings.
February 7, 2025 at 5:01 PM
Positions:
* Full-stack
* Research Engineer
* Research Scientist
* Systems Infrastructure Engineer
* Research intern
Feel free to reach out but chances are I will see your application if you apply online. I will post details on my internship later, but there are more openings.
* Full-stack
* Research Engineer
* Research Scientist
* Systems Infrastructure Engineer
* Research intern
Feel free to reach out but chances are I will see your application if you apply online. I will post details on my internship later, but there are more openings.
All investigations of faithfulness show that explanations' faithfulness is by default model and task-dependent. However, this is not the case when using FMMs. Thus, presenting a new paradigm for how to provide and ensure faithful explanations.
November 28, 2024 at 2:02 PM
All investigations of faithfulness show that explanations' faithfulness is by default model and task-dependent. However, this is not the case when using FMMs. Thus, presenting a new paradigm for how to provide and ensure faithful explanations.
FMMs are when models are designed such that measuring faithfulness is cheap and precise, which makes it possible to optimize explanations toward maximum faithfulness.
November 28, 2024 at 2:02 PM
FMMs are when models are designed such that measuring faithfulness is cheap and precise, which makes it possible to optimize explanations toward maximum faithfulness.
Self-explanations are when LLMs explain themselves. Current models are not capable of this, but we suggest how that could be changed.Diagram of self-explanations. Showing input going in, then the regular output and explanation going out.
November 28, 2024 at 2:02 PM
Self-explanations are when LLMs explain themselves. Current models are not capable of this, but we suggest how that could be changed.Diagram of self-explanations. Showing input going in, then the regular output and explanation going out.
We ask the question: How to provide and ensure faithful explanations for general-purpose NLP models? The main thesis is that we should develop new paradigms in interpretability. The two new paradigms explored are faithfulness measurable models (FMMs) and self-explanations.
November 28, 2024 at 2:02 PM
We ask the question: How to provide and ensure faithful explanations for general-purpose NLP models? The main thesis is that we should develop new paradigms in interpretability. The two new paradigms explored are faithfulness measurable models (FMMs) and self-explanations.
The full thesis is available at arxiv.org/abs/2411.17992. Thanks to @sivareddyg.bsky.social and @sarath-chandar.bsky.social for supervising me throughout all these years. It's been a great journey and I'm very grateful for their support.
New Faithfulness-Centric Interpretability Paradigms for Natural Language Processing
As machine learning becomes more widespread and is used in more critical applications, it's important to provide explanations for these models, to prevent unintended behavior. Unfortunately, many curr...
arxiv.org
November 28, 2024 at 2:02 PM
The full thesis is available at arxiv.org/abs/2411.17992. Thanks to @sivareddyg.bsky.social and @sarath-chandar.bsky.social for supervising me throughout all these years. It's been a great journey and I'm very grateful for their support.
Hi, can you add me thanks 🙂
November 27, 2024 at 4:13 PM
Hi, can you add me thanks 🙂